The treatment of many human epileptic syndromes remains unsatisfactory. While anticonvulsant medications allow about 75% of epileptics to achieve excellent seizure control, the remaining 25% of patients suffer from a combination of continued seizures and medication toxicity. It is unlikely that a single medical breakthrough will provide a cure for all of these refractory patients. Focal neocortical epilepsies have proven particularly difficult to manage. While some respond to anticonvulsants, a large fraction remains intractable to medical therapy. This group can respond to cortical resection, but surgical management is problematic. Exact identification of the epileptogenic focus can be complicated and there is a risk of unanticipated, irreversible neurological deficits after resection. Focal cortical cooling has the potential to improve the evaluation and treatment of this epileptic subgroup. The aims of the experiments described in this application are to investigate the potential of focal cooling with thermoelectric (Peltier) chips to rapidly terminate chronic seizure discharges, determine the degree of cooling required to stop these seizures, determine whether cooling can prevent seizures, and develop computer programs that recognize and anticipate seizures in "real time". In addition, the potential pathological consequences of cortical cooling will be determined. These experiments represent a necessary first step toward utilizing these techniques for the therapy of human epilepsy. These experiments will utilize models of acute and chronic rodent neocortical seizures and small Peltier devices developed for the microelectronics industry. If Peltier devices can control focal seizures in our models, they will be refined for future experiments to investigate their potential role in mapping and controlling epileptogenic neocortex in man. [unreadable] [unreadable]